Send email for updates
Parhar K.K., Wormith J.S, Derkzen D.M. et al.
Criminal Justice and Behavior: 2008, 35(9), p. 1109–1135
Unable to obtain a copy by clicking title? Try asking the author for a reprint by adapting this prepared e-mail or by writing to Dr Parhar at firstname.lastname@example.org. You could also try this alternative source.
This comprehensive synthesis of 129 studies of offender treatment for problems such as substance use found increasing treatment impact as the degree to which the offender was free to choose the treatment increased. At the bottom end, mandated treatment in custody appeared a waste of time and money.
Summary It is well established that correctional treatment programmes can reduce recidivism. Possibly for this reason, and possibly also because it is thought that few offenders would voluntarily enter treatment, legally mandated treatment has become increasingly common as a way to safeguard the community against crime. However, its effectiveness is commonly queried by clinicians who contend treatment can only work if participants choose to attend because of their own internal motivation.
The featured review aimed to establish the truth of this contention. 'Working' was operationalised as crime reduction as indicated most often by arrest rates. However, assessing the degree to which offenders were forced in versus chose to enter treatment was more complicated. Few studies directly assessed this dimension. Nevertheless, the legal context within which treatment is offered and associated incentives and sanctions may stand as a proxy for choice versus force – and this context often is described.
The degree to which offenders had been legally coerced in to treatment was ranked on a five-point scale:
1 Mandated involuntary: treatment must be attended if offenders are to avoid (continued) imprisonment or other negative consequences and/or court treatment has been ordered by the court as part of their sentence.
2 Mandated coerced: offenders are mandated to treatment but sanctions for not complying are in practice minimal, nil or inconsistently applied.
3 Non-mandated coerced with legal consequences: though not mandated to attend or complete treatment, offenders face some legal consequences if they do not such as being sent back to court.
4 Non-mandated coerced: there is no treatment mandate but offenders may receive incentives if they attend treatment, such as early release from prison to non-custodial supervisions.
5 Non-mandated or voluntary: offenders freely volunteer to attend treatment with no external sanctions or benefits.
These five levels were also collapsed in to mandated (level 1), coerced (levels 2 to 4), or voluntary (level 5).
The treatments offered were also assessed for quality in terms of their adherence to the principles of effective correctional treatment encapsulated in the risk, need, and responsivity model, but in practice this variable was unrelated to outcomes.
Relevant studies were sought which assessed recidivism, enabled an assessment of the pressure to enter treatment, and compared outcomes for treated offenders to those not treated by the programme being evaluated. 129 such studies were found published between 1970 and 2005, all but a few conducted in the United States (81%) or Canada (15%). In nearly 4 in 10 of the studies, the treatment addressed substance use; sexual offending, violence and juvenile delinquency were among the other issues addressed.
Assessed on the five-point scale, there was a significant positive relationship between the degree to which the treatment was voluntary and the strength of its association with reduced recidivism expressed an effect size. A standard way of expressing the magnitude of a difference (eg, between outcomes in control and intervention groups) applicable to most quantitative data. Enables different measures taken in different studies to be compared or (in meta-analyses) combined. Based on expressing the difference in the average outcomes between control and experimental groups as a proportion of how much the outcome varies across both groups. The most common statistic used to quantify this difference is called Cohen's d. Conventionally this is considered to indicate a small effect when no greater than 0.2, a medium effect when around 0.5, and a large effect when at least 0.8. This was true when recidivism was measured as any known or likely return to crime in general (most commonly arrests), when the effect size was a small to medium 0.24. It was also the case (effect size 0.21) when other studies which only reported on the specific type of offence for which the offender was being treated were also added to the pool of analysed studies (termed 'any' recidivism). However, when the focus was sharpened to the 52 studies which reported recidivism specific to the problem being treated, the effect size (0.13) was smaller and no longer statistically significant.
Whether the offender being treated was in custody significantly affected recidivism outcomes from the treatment, so the analyses were repeated separately for treatments for offenders in custody versus not in custody. Again, voluntariness was not significantly related to specific recidivism but was to recidivism more generally. In custody the significant effect sizes were 0.43 for general recidivism and 0.41 when specific recidivism was also included in the calculations. Outside custody, only when general and specific recidivism were pooled was the effect size (0.24) statistically significant. Though not significant, the other effect sizes at around 0.20 were not negligible.
The contention that mandated treatment extends treatment retention was tested by relating voluntariness to the proportion of offenders who dropped out of treatment. Across all 56 relevant studies, the reverse was found: as the degree of voluntariness rose the drop-out rate significantly declined. Analysed separately, this was no longer the case for offenders in versus out of custody. Also the degree to which the treatments reduced recidivism was not related to how well they had retained the offenders.
When treatments were categorised as either mandated, coerced , or voluntary, in each category the treatments significantly or near -significantly reduced recidivism. But the treatment impacts in respect of general and any recidivism were greatest for voluntary treatment (significantly so in comparison with mandated treatment), and least when treatment had been mandated. In other words, and confirming the five-point scale analysis, the more the treatments had been coerced, the less their impact. When the focus was narrowed to offenders in prison, only voluntary treatment was significantly associated with crime reductions. Outside prison all three categories of treatments were associated with significantly reduced recidivism.
Rather than simply categorising treatment as mandated or voluntary, the featured review attempted to create a more precise measure reflecting various degrees of coercion. Outside custody, voluntary, coerced and mandated treatment were all significantly associated with crime reductions and in that rank order of magnitude from largest to smallest. The rank order was the same in custodial settings, but there neither coerced nor mandated treatment were significantly associated with crime reductions. Across both settings, the strongest effect sizes were found for voluntary treatment (without coercion) and the weakest for mandated treatment. These findings suggest the common use of mandate and coercion to promote treatment entry and retention in the criminal justice system may adversely affect treatment outcome.
If offenders are required by courts to attend treatment in custody, the findings suggest that in terms of crime prevention, the treatment is likely to be a waste of time and money. However, a degree of coercion in the form of incentives and legal consequences related to entering treatment does not seem unduly damaging either in prison or the community.
However, these results derive not from 'head-to-head' comparisons when offenders have been randomly allocated to mandated, coerced or voluntary treatment. Instead they take the form of an association between the degree of legal pressure on the offender to accept treatment and the degree to which that treatment is followed by reductions in offending. What causes this association is unclear because this methodology cannot eliminate all the other influences which might have led to the result other than the degree of coercion. Among those other influences may be the act of volunteering for treatment, differences between patients such as the nature and strength of their motivation to enter and succeed in treatment, or differences between coerced and voluntary treatments such as in their intensity. Perhaps too treatment staff are more responsive to clients who have volunteered.
commentary Since these findings do not derive from 'head-to-head' comparisons, the possibility cannot be excluded that what looks like a gradient of increased treatment impact as it becomes more voluntary, is instead a gradient of selectivity in the offenders who enter treatment. Motivation for treatment and the desire to make the most of it can be expected to be greater among offenders who have freely chosen (even if in response to pressure from family or employers or other sources) to seek treatment rather than those who have effectively been forced. In this case, voluntary treatment reaps the benefits of a more motivated caseload which accounts for its apparent relative effectiveness, while mandated treatment has to take all comers regardless of motivation.
It also seems possible that what seems a gradient of effectiveness is at least partly a gradient of severity. Like patients in general, offenders are likely to voluntarily choose treatment when their problems have become intolerable to them and/or to their families and other associates. In contrast, to avoid sanctions, coerced and mandated offenders may enter treatment even if neither they nor their close social circle feel it is warranted by their problems. In this scenario, volunteer treatment entrants both have more scope to improve and more reason to do so.
Despite these caveats, the analysis offers a corrective to the common assumption based on a more limited set of studies that coerced and mandated treatment are as effective as voluntary. If the findings do indeed show that voluntary treatment is considerably more effective in reducing crime, instead of extending the net of mandated treatment, the task becomes to maximise the attractiveness and availability of treatment so that a higher proportion of offenders chose this rather than having to be forced in to it.
Last revised 14 February 2012. First uploaded
Give us your feedback on the site (two-minute survey)
Open Effectiveness Bank home page
Add your name to the mailing list to be alerted to new studies and other site updates
HOT TOPIC 2016 Don’t treat, just test and sanction
DOCUMENT 2012 Will intensive testing and sanctions displace treatment?
REVIEW 2011 A new paradigm for long-term recovery